Apply JuliaFormatter to fix code formatting

ext/FastPowerEnzymeExt.jl

@@ -13,25 +13,27 @@ function Enzyme.EnzymeRules.forward(config::FwdConfig,
     y = _y.val
     ret = func.val(x, y)
     T = typeof(ret)
-    if !(_x isa Const) 
-        dxval = _x.dval * y * (fastpower(x,y - 1))  
-    else 
+    if !(_x isa Const)
+        dxval = _x.dval * y * (fastpower(x, y - 1))
+    else
         dxval = make_zero(_x.val)
     end
-    if !(_y isa Const)  
-        dyval = x isa Real && x<=0 ? Base.oftype(float(x), NaN) : _y.dval*(fastpower(x,y))*log(x)
-    else 
+    if !(_y isa Const)
+        dyval = x isa Real && x<=0 ? Base.oftype(float(x), NaN) :
+                _y.dval*(fastpower(x, y))*log(x)
+    else
         dyval = make_zero(_y.val)
-    end 
+    end
     if RT <: DuplicatedNoNeed
-        return convert(T,dxval + dyval)
+        return convert(T, dxval + dyval)
     else
         return Duplicated(ret, convert(T, dxval + dyval))
     end
 end
 
-function EnzymeRules.augmented_primal(config::Enzyme.EnzymeRules.RevConfigWidth{1}, 
-        func::Const{typeof(fastpower)}, ::Union{Type{<:Active}, Type{<:Const}}, x::Union{Const,Active}, y::Union{Const,Active})
+function EnzymeRules.augmented_primal(config::Enzyme.EnzymeRules.RevConfigWidth{1},
+        func::Const{typeof(fastpower)}, ::Union{Type{<:Active}, Type{<:Const}},
+        x::Union{Const, Active}, y::Union{Const, Active})
     if EnzymeRules.needs_primal(config)
         primal = func.val(x.val, y.val)
     else
@@ -40,12 +42,15 @@ function EnzymeRules.augmented_primal(config::Enzyme.EnzymeRules.RevConfigWidth{
     return EnzymeRules.AugmentedReturn(primal, nothing, nothing)
 end
 
-function EnzymeRules.reverse(config::Enzyme.EnzymeRules.RevConfigWidth{1}, 
-        func::Const{typeof(fastpower)}, dret, tape, _x::Union{Const,Active}, _y::Union{Const,Active})
+function EnzymeRules.reverse(config::Enzyme.EnzymeRules.RevConfigWidth{1},
+        func::Const{typeof(fastpower)}, dret, tape, _x::Union{Const, Active}, _y::Union{
+            Const, Active})
     x = _x.val
     y = _y.val
-    dxval =  _x isa Const ? nothing : dret.val * y * (fastpower(x,y - 1))
-    dyval = _y isa Const ? nothing : (x isa Real && x<=0 ? Base.oftype(float(x), NaN) : dret.val * (fastpower(x,y))*log(x))
+    dxval = _x isa Const ? nothing : dret.val * y * (fastpower(x, y - 1))
+    dyval = _y isa Const ? nothing :
+            (x isa Real && x<=0 ? Base.oftype(float(x), NaN) :
+             dret.val * (fastpower(x, y)) * log(x))
     return (dxval, dyval)
 end
 

ext/FastPowerMooncakeExt.jl

@@ -1,6 +1,6 @@
 module FastPowerMooncakeExt
 
 using FastPower, Mooncake
-Mooncake.@mooncake_overlay FastPower.fastpower(x,y) = x^y
+Mooncake.@mooncake_overlay FastPower.fastpower(x, y) = x^y
 
-end
+end

src/FastPower.jl

@@ -11,7 +11,7 @@ module FastPower
     c = 0.523692f0
     # IEEE is sgn(1):exp(8):frac(23) representing
     # (1+frac)*2^(exp-127). 1+frac is called the significand
-    
+
     # get exponent
     ux1i = reinterpret(UInt32, x)
     exp = Int32((ux1i & 0x7F800000) >> 23)
@@ -24,7 +24,7 @@ module FastPower
         ux2i = (ux1i & 0x007FFFFF) | 0x3f000000
         exp -= 0x7e # 126 instead of 127 compensates for division by 2
     end
-    signif = reinterpret(Float32, ux2i) 
+    signif = reinterpret(Float32, ux2i)
     quot = muladd(signif, a, b) / (signif + c)
     return muladd(signif - 1.0f0, quot, exp)
 end

test/runtests.jl

@@ -27,22 +27,23 @@ end
     @testset for RT in (Duplicated, DuplicatedNoNeed),
         Tx in (Const, Duplicated),
         Ty in (Const, Duplicated)
-
         x = 1.0
         y = 0.5
-        test_forward(fastpower, RT, (x, Tx), (y, Ty), atol = 1e-4, rtol=1e-3)
+        test_forward(fastpower, RT, (x, Tx), (y, Ty), atol = 1e-4, rtol = 1e-3)
     end
 end
 
 @testset "Fast pow - Enzyme reverse rule" begin
     @testset for RT in (Active,), Tx in (Active, Const), Ty in (Active, Const)
         x = 1.0
         y = 0.5
-        test_reverse(fastpower, RT, (x, Tx), (y, Ty), atol = 1e-4, rtol=1e-3)
+        test_reverse(fastpower, RT, (x, Tx), (y, Ty), atol = 1e-4, rtol = 1e-3)
     end
 end
 
-mooncake_derivative(f,x) = Mooncake.value_and_gradient!!(Mooncake.build_rrule(f, x), f, x)[2][2]
+function mooncake_derivative(f, x)
+    Mooncake.value_and_gradient!!(Mooncake.build_rrule(f, x), f, x)[2][2]
+end
 @testset "Fast pow - Other AD Engines" begin
     x = 1.5123233245141
     y = 0.22352354326